_ Git - youtube-dl/blob - youtube_dl/extractor/yandexmusic.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import hashlib
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_str
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     float_or_none,
  13     sanitized_Request,
  14     urlencode_postdata,
  15 )
  16
  17
  18 class YandexMusicBaseIE(InfoExtractor):
  19     @staticmethod
  20     def _handle_error(response):
  21         error = response.get('error')
  22         if error:
  23             raise ExtractorError(error, expected=True)
  24
  25     def _download_json(self, *args, **kwargs):
  26         response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
  27         self._handle_error(response)
  28         return response
  29
  30
  31 class YandexMusicTrackIE(YandexMusicBaseIE):
  32     IE_NAME = 'yandexmusic:track'
  33     IE_DESC = 'Яндекс.Музыка - Трек'
  34     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
  35
  36     _TEST = {
  37         'url': 'http://music.yandex.ru/album/540508/track/4878838',
  38         'md5': 'f496818aa2f60b6c0062980d2e00dc20',
  39         'info_dict': {
  40             'id': '4878838',
  41             'ext': 'mp3',
  42             'title': 'Carlo Ambrosio - Gypsy Eyes 1',
  43             'filesize': 4628061,
  44             'duration': 193.04,
  45         }
  46     }
  47
  48     def _get_track_url(self, storage_dir, track_id):
  49         data = self._download_json(
  50             'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
  51             % storage_dir,
  52             track_id, 'Downloading track location JSON')
  53
  54         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
  55         storage = storage_dir.split('.')
  56
  57         return ('http://%s/get-mp3/%s/%s?track-id=%s&from=service-10-track&similarities-experiment=default'
  58                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
  59
  60     def _get_track_info(self, track):
  61         thumbnail = None
  62         cover_uri = track.get('albums', [{}])[0].get('coverUri')
  63         if cover_uri:
  64             thumbnail = cover_uri.replace('%%', 'orig')
  65             if not thumbnail.startswith('http'):
  66                 thumbnail = 'http://' + thumbnail
  67         return {
  68             'id': track['id'],
  69             'ext': 'mp3',
  70             'url': self._get_track_url(track['storageDir'], track['id']),
  71             'title': '%s - %s' % (track['artists'][0]['name'], track['title']),
  72             'filesize': int_or_none(track.get('fileSize')),
  73             'duration': float_or_none(track.get('durationMs'), 1000),
  74             'thumbnail': thumbnail,
  75         }
  76
  77     def _real_extract(self, url):
  78         mobj = re.match(self._VALID_URL, url)
  79         album_id, track_id = mobj.group('album_id'), mobj.group('id')
  80
  81         track = self._download_json(
  82             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
  83             track_id, 'Downloading track JSON')['track']
  84
  85         return self._get_track_info(track)
  86
  87
  88 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
  89     def _build_playlist(self, tracks):
  90         return [
  91             self.url_result(
  92                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
  93             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
  94
  95
  96 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
  97     IE_NAME = 'yandexmusic:album'
  98     IE_DESC = 'Яндекс.Музыка - Альбом'
  99     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
 100
 101     _TEST = {
 102         'url': 'http://music.yandex.ru/album/540508',
 103         'info_dict': {
 104             'id': '540508',
 105             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
 106         },
 107         'playlist_count': 50,
 108     }
 109
 110     def _real_extract(self, url):
 111         album_id = self._match_id(url)
 112
 113         album = self._download_json(
 114             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
 115             album_id, 'Downloading album JSON')
 116
 117         entries = self._build_playlist(album['volumes'][0])
 118
 119         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 120         year = album.get('year')
 121         if year:
 122             title += ' (%s)' % year
 123
 124         return self.playlist_result(entries, compat_str(album['id']), title)
 125
 126
 127 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 128     IE_NAME = 'yandexmusic:playlist'
 129     IE_DESC = 'Яндекс.Музыка - Плейлист'
 130     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)'
 131
 132     _TESTS = [{
 133         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 134         'info_dict': {
 135             'id': '1245',
 136             'title': 'Что слушают Enter Shikari',
 137             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 138         },
 139         'playlist_count': 6,
 140     }, {
 141         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 142         # https://github.com/rg3/youtube-dl/issues/6666)
 143         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 144         'info_dict': {
 145             'id': '1036',
 146             'title': 'Музыка 90-х',
 147         },
 148         'playlist_count': 310,
 149     }]
 150
 151     def _real_extract(self, url):
 152         playlist_id = self._match_id(url)
 153
 154         webpage = self._download_webpage(url, playlist_id)
 155
 156         mu = self._parse_json(
 157             self._search_regex(
 158                 r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'),
 159             playlist_id)
 160
 161         playlist = mu['pageData']['playlist']
 162         tracks, track_ids = playlist['tracks'], playlist['trackIds']
 163
 164         # tracks dictionary shipped with webpage is limited to 150 tracks,
 165         # missing tracks should be retrieved manually.
 166         if len(tracks) < len(track_ids):
 167             present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
 168             missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
 169             request = sanitized_Request(
 170                 'https://music.yandex.ru/handlers/track-entries.jsx',
 171                 urlencode_postdata({
 172                     'entries': ','.join(missing_track_ids),
 173                     'lang': mu.get('settings', {}).get('lang', 'en'),
 174                     'external-domain': 'music.yandex.ru',
 175                     'overembed': 'false',
 176                     'sign': mu.get('authData', {}).get('user', {}).get('sign'),
 177                     'strict': 'true',
 178                 }))
 179             request.add_header('Referer', url)
 180             request.add_header('X-Requested-With', 'XMLHttpRequest')
 181
 182             missing_tracks = self._download_json(
 183                 request, playlist_id, 'Downloading missing tracks JSON', fatal=False)
 184             if missing_tracks:
 185                 tracks.extend(missing_tracks)
 186
 187         return self.playlist_result(
 188             self._build_playlist(tracks),
 189             compat_str(playlist_id),
 190             playlist['title'], playlist.get('description'))